Objective: Biliary atresia (BA) represents the most frequent cause of liver transplantation (LT) in the pediatric population. An association between BA and portal hypoplasia has been previously described, making this group at higher risk of portal complications, which could eventually lead to portal hypertension, graft failure or even graft loss. The main objective of this work is to analyze and compare the results before and after the implementation of longitudinal portoplasty (LP) as the technique of choice during LT for patients with BA and hypoplasic portal vein in our center.

Materials and Methods: Retrospective study including all patients transplanted due to BA and a preoperative diagnosis of portal hypoplasia (diameter measured by doppler ultrasound or CT scan less or equal to 5 mm). Results were analyzed before (group 1) and after the introduction of the new reconstruction technique in 2016 (group 2). Primary endpoints were the occurrence of early portal vein thrombosis (within 30 days of transplantation) and patient survival. Secondary endpoints were the occurrence of portal complications.

Results: A total of 100 patients (out of 346 transplants, 78.1% of patients with BA) had a preoperative diagnosis of BA with portal hypoplasia and were included (50 patients in each group). Baseline characteristics did not differ among groups. Reduced grafts were used in 96% of the cases, and living donors in 43%. Overall rate of portal early thrombosis was 22% (n=11) in the first group and 2% (n=2) in the second (p= 0.002). An association between early portal early thrombosis and LP was found (p: 0.025 OR: 0.089 95% CI: 0.011-0.74). With regard to the technique used, portal early thrombosis was found in 12.7% (n=6) of end to end anastomosis, in 41.7% (n=5) of graft interposition and in 2.4% (n=1) of LP (p: 0.001). Patient survival at 12 and 24 months in the first group was 79.5% and 77.3%, while in the second group was 92% and 92% (log rank test 0.048). Median follow-up was 2 years. No statistical significant difference between survival and LP was found (p: 0.06 HR: 0.34 95% CI: 0.11-1.05). Overall rate of portal complications was 28%. With the implementation of LP there was a decrease of 24% in portal complications (p: 0.008). No statistical difference was found for stenosis among groups.

Conclusion: Since the implementation of LP, a decrease in early portal thrombosis was seen. Limitations of this study are the fact that it is a retrospective and single institution study. Although results showed favorable trends towards LP, survival did not reach statistical significance yet, possibly due to lack of power in this context of limited cases to analyze. Larger sample size and longer follow-up studies are required to confirm these findings in the future. In conclusion, the routine use of LP could possibly minimize the occurrence of portal complications in patients with BA and portal hypoplasia, and could be considered in these cases with good results.